Am Fam Physician. 2009;80(1):86-91
Background: Communicating health risk information to patients can be challenging. Information about the risk of death from specific medical conditions, such as cervical or prostate cancer, is often provided without context. In 2002, Woloshin and colleagues published a set of charts that used nationally representative mortality data to present comparative information on a patient's 10-year risk of death from various causes by age, sex, and smoking status. The authors subsequently updated and expanded these charts using improved methodology and more recent data sources. Tables 1 and 2 are the risk charts for men and women (current and never smokers).
| Age | Smoking status | Vascular Disease | Cancer | Infection | Lung disease | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Heart disease | Stroke | Lung | Colon | Prostate | Pneumonia | Flu | AIDS | COPD | Accidents | All causes combined | ||
| 35 | Never smoker | 1 | 1 | <1 | <1 | <1 | <1 | <1 | 2 | <1 | 5 | 15 |
| Smoker | 7 | 1 | 1 | <1 | <1 | <1 | <1 | 2 | <1 | 5 | 42 | |
| 40 | Never smoker | 3 | 1 | 1 | 1 | <1 | <1 | <1 | 2 | <1 | 6 | 24 |
| Smoker | 14 | 2 | 4 | 1 | <1 | <1 | <1 | 2 | 1 | 6 | 62 | |
| 45 | Never smoker | 6 | 1 | 1 | 1 | <1 | <1 | <1 | 2 | <1 | 6 | 35 |
| Smoker | 21 | 3 | 8 | 1 | <1 | 1 | <1 | 2 | 1 | 6 | 91 | |
| 50 | Never smoker | 11 | 1 | 1 | 2 | 1 | 1 | <1 | 1 | <1 | 5 | 49 |
| Smoker | 29 | 5 | 18 | 2 | 1 | 1 | <1 | 1 | 3 | 5 | 128 | |
| 55 | Never smoker | 19 | 3 | 1 | 3 | 2 | 1 | <1 | 1 | 1 | 5 | 71 |
| Smoker | 41 | 7 | 34 | 3 | 1 | 2 | <1 | 1 | 7 | 4 | 178 | |
| 60 | Never smoker | 32 | 5 | 2 | 5 | 3 | 2 | <1 | 1 | 1 | 5 | 115 |
| Smoker | 56 | 11 | 59 | 5 | 3 | 3 | <1 | 1 | 16 | 4 | 256 | |
| 65 | Never smoker | 52 | 9 | 4 | 8 | 6 | 3 | <1 | <1 | 3 | 6 | 176 |
| Smoker | 74 | 16 | 89 | 7 | 6 | 5 | <1 | <1 | 26 | 5 | 365 | |
| 70 | Never smoker | 87 | 18 | 6 | 10 | 12 | 6 | <1 | <1 | 5 | 7 | 291 |
| Smoker | 100 | 26 | 113 | 9 | 10 | 9 | <1 | <1 | 45 | 6 | 511 | |
| 75 | Never smoker | 137 | 32 | 8 | 13 | 19 | 11 | <1 | <1 | 6 | 11 | 449 |
| Smoker | 140 | 39 | 109 | 11 | 15 | 16 | <1 | <1 | 60 | 9 | 667 | |
| Age | Smoking status | Vascular disease | Cancer | Infection | Lung disease | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Heart disease | Stroke | Lung | Breast | Colon | Ovarian | Cervical | Pneumonia | Flu | AIDS | COPD | Accidents | All causes combined | ||
| 35 | Never smoker | 1 | <1 | <1 | 1 | <1 | <1 | <1 | <1 | <1 | 1 | <1 | 2 | 14 |
| Smoker | 1 | 1 | 1 | 1 | <1 | <1 | <1 | <1 | <1 | 1 | <1 | 2 | 14 | |
| 40 | Never smoker | 1 | <1 | <1 | 2 | 1 | <1 | <1 | <1 | <1 | 1 | <1 | 2 | 19 |
| Smoker | 4 | 2 | 4 | 2 | <1 | <1 | <1 | <1 | <1 | 1 | 1 | 2 | 27 | |
| 45 | Never smoker | 2 | 1 | 1 | 3 | 1 | 1 | <1 | <1 | <1 | 1 | <1 | 2 | 25 |
| Smoker | 9 | 3 | 7 | 3 | 1 | 1 | <1 | <1 | <1 | 1 | 2 | 2 | 45 | |
| 50 | Never smoker | 4 | 1 | 1 | 4 | 1 | 1 | <1 | <1 | <1 | <1 | <1 | 2 | 37 |
| Smoker | 13 | 5 | 14 | 4 | 1 | 1 | <1 | <1 | <1 | <1 | 4 | 2 | 69 | |
| 55 | Never smoker | 8 | 2 | 2 | 6 | 2 | 2 | 1 | 1 | <1 | <1 | 1 | 2 | 55 |
| Smoker | 20 | 6 | 26 | 5 | 2 | 2 | 1 | 1 | <1 | <1 | 9 | 2 | 110 | |
| 60 | Never smoker | 14 | 4 | 3 | 7 | 3 | 3 | 1 | 1 | <1 | <1 | 2 | 2 | 84 |
| Smoker | 31 | 8 | 41 | 6 | 3 | 3 | 1 | 2 | <1 | <1 | 18 | 2 | 167 | |
| 65 | Never smoker | 25 | 7 | 5 | 8 | 5 | 4 | 1 | 2 | <1 | <1 | 3 | 3 | 131 |
| Smoker | 45 | 15 | 55 | 7 | 5 | 3 | 1 | 4 | <1 | <1 | 31 | 3 | 241 | |
| 70 | Never smoker | 46 | 14 | 7 | 9 | 7 | 4 | 1 | 4 | <1 | <1 | 5 | 4 | 207 |
| Smoker | 56 | 25 | 61 | 8 | 6 | 4 | 1 | 7 | <1 | <1 | 44 | 4 | 335 | |
| 75 | Never smoker | 89 | 30 | 7 | 11 | 10 | 5 | 1 | 8 | <1 | <1 | 6 | 7 | 335 |
| Smoker | 99 | 34 | 58 | 10 | 9 | 4 | <1 | 14 | <1 | <1 | 61 | 7 | 463 | |
The Study: Data sources for this study included the National Center for Health Statistics, the U.S. Census Bureau, and the American Cancer Society's Cancer Prevention Study II (a cohort study of 1.2 million adults that began in 1982). Risk charts were developed for common and less common causes of death in men and women (10 causes for men, 12 for women). The number of causes examined differed because of sex-specific causes of death (prostate, breast, cervical, and ovarian cancers). To calculate death rates, the authors divided the number of deaths in a particular age and sex group by the corresponding age and sex-specific population in the reference year (2004). Relative risks of death based on smoking status were obtained from the Cancer Prevention Study II. Higher death rates for current and former smokers were calculated by multiplying these relative risks by the baseline risk of never smokers. All death rates were presented as a number per 1,000 persons over 10 years.
Results: The risk charts illustrate several comparisons that may assist physicians in encouraging healthy behaviors and prioritizing preventive measures. Smoking dramatically increases the risk of death from vascular disease, cancer, and lung disease in men and women throughout the adult lifespan. At 75 years of age, a male smoker is more than 10 times as likely to die from lung cancer or chronic obstructive pulmonary disease (COPD) in the next 10 years than a never smoker. The strong association of smoking with death from certain causes leads to a statistical artifact in that smokers have lower death rates for a few conditions (colon cancer in men older than 65 years, breast cancer in women 55 to 70 years of age) than never smokers. However, the apparent “protective” effect from smoking actually results from more smokers dying earlier from competing causes of death (for example, a smoker who dies of COPD will not live long enough to die of colon cancer). Regardless of smoking status, the risk of death from vascular disease at any age far exceeds the risk of death from prostate or breast cancer.
Conclusion: The authors conclude that these risk charts provide reliable estimates of the magnitude of major health risks for adults in an understandable context. They caution that the death rates represent national averages and are not intended to provide more personalized risk estimates that account for the influence of individual genetic, behavioral, or environmental factors.